Skip to main content

Improving trust in state and local government

Insights from data

State and local governments are on the front lines of delivering services—and of restoring trust in government. New survey data provides insight into what is driving perceptions of trust in state and local government functions



Democracy thrives on trust. In the United States, trust in government has been battered.

The data is clear: The federal government is experiencing big problems with trust. In April 2021, trust in the federal government stood at only 24%—a slight increase from 2020 (21%), but still near-historic lows.1 Many citizens are concerned with the federal government’s ability to fairly and effectively implement public policy (see Rebuilding trust in government). Some corporate CEOs seem concerned as well. In a January 2021 survey, corporate CEOs wanted restoring trust in government to be the new administration’s top priority—above infrastructure, economic recovery, and tax policy.2

The federal government is clearly facing a trust deficit, but what about trust in state and local governments? To find out, Deloitte surveyed 6,152 adults across the United States. (See sidebar, “Understanding trust” for details on the survey methodology.)

The bottom line: State and local governments, in general, were viewed more favourably than the federal government. Our survey, however, uncovered a number of areas where trust appears to be lacking. Moreover, the survey goes deeper into why citizens trust or tend to not trust various government entities.


Three key themes emerged from the survey findings:

  • The more local, the more trusted: Trust in government varies dramatically at different levels of government.
  • Digital experience drives trust: A citizen’s digital experience with a government agency was a strong predictor of their overall level of trust. 
  • Mission matters: The mission of various government agencies can greatly influence citizens’ perception of trust. 

The survey findings also offer insights on what elements of trust—competence or intent—respondents said agencies were doing well or poorly in. 

As we look at these differences, it is important to keep in mind that our trust scores do not precisely “measure” trust, which is impossible. Like happiness, trust is an abstract concept that cannot be measured but can only be assessed. In the survey, we asked respondents about their perceptions around the various drivers of trust and quantify their inferred trust levels based on their answers.

Understanding trust

How can you determine how trusted an organisation is? To answer that question, Deloitte developed a TrustID framework that is designed to get to the heart of trust. At its core, trust can be assessed by two elements—perceived competence and positive intent. For example, even if your dentist is very skilled, they won’t earn your trust if they recommend work you don’t need. On the flip side, a dentist who cares deeply for their patients but can’t fill a cavity shouldn’t be trusted either.  

Digging a little deeper, a government’s operational capability and reliability demonstrate its competence, while its humanity and transparency demonstrate positive intent. In practice, assessing trust can be very tricky, as it reflects the perception of an individual, not the reality of the organisation’s conduct. Trust is the combination of beliefs, perceptions, and feelings, shaped by our individual and collective experiences. Moreover, trust can sometimes be misplaced, as we deem a person or organisation to be doing a good job when they are not—or vice versa. In general, however, trust is often built over time by an organisation performing in a manner that demonstrates a high level of competence and an empathetic intent (figure 1). (For more details on Deloitte’s TrustID framework, please see Rebuilding trust in government.)

Using this framework, Deloitte surveyed 6,152 Americans in April 2021 to understand their levels of trust in the US state and local governments. We asked citizens about their views on how 16 different government agencies and departments across 41 states performed on four trust signals: humanity, transparency, capability, and reliability. 

Each trust signal and the composite trust score (average of four trust signal scores) can be measured on a scale of -100 to +100. Each survey response falls into one of the three distinct categories: low trust, mid-level trust, and high trust. The final trust signal score is the difference between high-trust and low-trust responses. The highest possible trust score is +100 (all the responses in the high-trust category), and the lowest possible trust score is -100 (all the responses in the low-trust category). (See the Appendix for more details.)

Show more

Key finding no. 1: Distant government tends to be distrusted government

People tend to trust their local government more than state government and trust the federal government least of all, according to respondents to our survey (figure 2).

This isn’t terribly shocking since the federal government is the most distant from citizens—it has fewer direct interactions and is less politically accessible. On the other hand, local government officials are closest to the people and more likely to hear directly from unhappy constituents. It is easier to go to your local town hall and access a town official or department head directly than it is to travel to Washington, DC in hopes of meeting someone in a position of authority. 

Decisions made by local government on various issues ranging from education and housing to public safety tend to directly impact people. More recently, state and local governments have been in the front line of COVID-19 response, managing the supply of necessary medical equipment as well as restoring the economy, which could be a major factor impacting present levels of trust in state and local governments.

Key finding no. 2: People trust individual agencies more than the overall government

Survey results indicate that respondents’ trust in overall government is less than the sum of its parts (figure 3). Based on the programmes we surveyed, respondents trusted individual agencies more than overall state government. For example, state agencies we surveyed received a 26 composite trust score versus a composite trust score of 6 for overall state government. This again points to the “perceptive” nature of trust, since trust in an agency might be based on interactions, but perceptions of the abstraction of “state government” are more likely to be associated with the politics of state government and influenced by media reports. This phenomenon also shows up in “Fenno’s paradox,” where people tend to dislike Congress but like their own congressperson.3

Across demographic groups, respondents score the agencies of government that they have interacted with higher on trust as compared to the whole of government (figure 4). Respondents in general indicated that they trust the parts of government more than the whole. There are some differences in how respondents of different ages view government: for example, individuals over 75 showed the highest trust levels of both state government in general and state agencies with which they had interacted. But the biggest variance, showing up at all ages and all incomes, is that respondents trust state agencies more than state government in the abstract.

Key finding no. 3: A citizen’s digital experience is a strong predictor of trust levels

Our survey revealed another interesting finding: individuals who are pleased with a state governments’ digital services also tend to rate the state highly in measures of overall trust. In general, respondents rate state and local agencies high on trust if they think that state governments’ digital services are easy to use, that governments’ web-based services help them accomplish what they need, and that the state government safeguards their data well.

On the flip side, citizens who are not satisfied with their state government’s digital services tended to generate lower trust scores (figure 5). For example, the trust score of those who disagreed with the statement “My state government’s internet and digital services are easy to use” was -13. These results suggest that improving the digital experience could be a key way to enhance overall trust.

As mentioned earlier, citizens tend to trust proximate government more than distant government. By design, digital services make distant services more proximate, creating a direct interaction. Because digital is now a first point of interaction for government to generate a positive impression, a positive online experience and secure and user-friendly services can be very important. 


Key finding no. 4: Unemployment Insurance, motor vehicles, and law enforcement are among the least trusted state agencies

Of the 16 agencies in our survey, respondents have the highest trust in child care services, housing assistance and food assistance programmes. The state agencies with the lowest trust scores are unemployment insurance agencies, departments of motor vehicles, and law enforcement agencies (figure 6).

One of the challenges of assessing trust in the public sector is the interplay between an agency’s mission and how “trusted” it is. In the abstract, we may intellectually support efforts to enforce traffic safety laws and restrictions on flying with dangerous items. In concrete terms, however, our personal experiences with receiving a speeding ticket from police or being forced to throw away shampoo at the airport may be more likely to generate negative feelings. In part, this is because “trust” reflects an individual’s perception and not an organisation’s actual performance. An agency providing benefits might create more positive feelings than agencies that enforce rules. For example, a courteously conducted IRS audit or traffic stop might still be inconvenient—which may put enforcement and regulatory agencies at a disadvantage in terms of trust scores. The mission paradox is that while nobody wants to be pulled over for speeding, everybody wants the other fellow who is speeding to get pulled over.

Deloitte’s retail-to-regulator (R2R) framework4 categorises government mission areas into six broad archetypes (figure 7). Depending on an agency’s mission and where it lands on the R2R spectrum can significantly influence how that agency is perceived. Moreover, understanding where an agency falls on the R2R spectrum can provide insight into an agency’s relationship with its customers and how it can improve its performance.

The top three agencies in our survey with the highest trust are all retailer-like agencies that provide aid to low-income families. Survey respondents considered these three agencies more capable, providing more reliable services, and at the same time, doing so with good intent compared to other agencies. 

Among the bottom three, Unemployment Insurance (UI) and DMVs also constitute the retailer-like archetype, while law enforcement agencies belong to the enforcer archetype in the R2R spectrum. 

Looking back at recent events can help us find possible reasons for respondents’ low trust in these agencies. 

  • For UI agencies, low trust is likely attributed to the struggles many Americans faced in accessing unemployment benefits during the COVID-19 pandemic. There were long delays for numerous unemployed workers, many of whom were badly in need of funds, as well as problems with fraud.5
  • State DMVs, too, have been strained by COVID-19, outdated computer systems, enormous backlogs, and increasing customer wait times, which could contribute to low trust among citizens.6
  • For law enforcement agencies, one factor for low trust could be recent media coverage of high-profile cases of police misconduct and subsequent protests. For the first time in the three decades that Gallup has been tracking it, public confidence in the police has fallen below 50%.7

Assessing the mission in which an agency fits can help identify the key areas that should be addressed to rebuild trust. For example, retailer-like agencies that directly deliver essential services to citizens can improve trust by providing a better customer experience—as with fast, efficient transactions. On the other hand, while likely a much greater challenge, driving trust in enforcement agencies could entail encouraging greater voluntary compliance.

Restoring trust

Restoring public trust is an important priority for state and local governments in general, as well as individual agencies. On the one hand, higher trust may often be a reflection that government is delivering competent and empathetic services. Higher trust in government may also help agencies to be more successful in their missions, as increased trust may lead to desirable behaviors such as greater compliance with regulators, participation in optional programmes, and engagement with government. Strong trust can enable greater government effectiveness, while low trust can increase costs and make it more difficult to achieve desired outcomes.

Our survey insights point to some key steps that state governments can take to restore trust:

1) Identify where trust is lacking: The first step to restoring trust is assessing trust levels among the population, and then digging deeper to find out what’s behind those views. Government could benefit from greater insight into the trust perception of its people. Polls or surveys can be valuable instruments for discovering which agencies are trusted, which are not, as well as what groups of people lack trust in public institutions and why. 

2) Focus on your agency’s digital experience: Our results showed a clear link between satisfaction with online services and trust in government. Keeping the end-to-end user experience in mind can help agencies design services that are easy to use and deliver positive results. Governments should consider enhancing the digital experience as a way to foster greater trust in government. 

3) Choose trust building actions based on agency mission: Assessing the appropriate archetype in which an agency fits based on its mission can help to identify the key trust signals—humanity, transparency, capability, or reliability—that an agency might focus on to restore trust. Benefit-providing agencies and enforcement agencies may gain from different approaches.

4) Proactively and transparently sharing performance data: Government agencies sometimes miss the opportunity to share good news about what they have accomplished. The shortcomings of government services tend to be highlighted, while success stories are often overlooked. Government agencies that transparently share their own results—good and bad—can help combat this bias.

5) Make the journey to restoring public trust a priority: The lack of trust in government has reached a point where it merits the attention of public leaders. The journey to restoring trust should include not only measuring it, but also enhancing services and managing trust perception in a way that ensures the areas of weaknesses are addressed. 


Understanding the trust signal and composite scores

The composite trust scores shared in the paper were calculated by grouping the responses to individual questions into three buckets: low trust, mid-level trust, and high trust, as shown in the figure below. We then subtracted the low-trust responses from the high-trust responses to generate a net score for that question, which could theoretically range from -100 to +100. The responses for each question were combined according to the drivers of trust, with humanity and transparency being grouped to yield a value for intent, and capability and reliability being grouped to yield a value for competence.

Trust in state and local government

It’s time to recognise the importance of trust in our public institutions. When trust in government declines, it becomes harder for state and local government to execute on essential tasks. Trust helps government connect with residents and encourages voluntary compliance with rules and regulations.

Trust can elevate state and local government, improving performance and connecting with the common good. Our survey shows what drives confidence in government, and which public agencies are trusted—and why. Now more than ever, state and local governments should be thinking about trust as an essential component of performance. Deloitte can help public officials measure, enhance, and amplify trust in state and local government agencies.

Learn more

  1. Pew Research Center, “Americans’ views of government: Low trust, but some positive performance ratings ,” September 14, 2020; Pew Research Center, “Public trust in government: 1958–2021 ,” May 17, 2021.

    View in Article
  2. Lance Lambert, “Will Biden make corporate America happy? Here’s CEOs’ to-do list for the new administration ,” Fortune , January 28, 2021.

    View in Article
  3. Harry Enten, “Disliking Congress, as a whole and as individuals ,” FiveThirtyEight, July 1, 2014; named after Richard Fenno who discussed this in his 1978 book Home Style: House Members in Their Districts .

    View in Article
  4. Deloitte, Retail to regulator: A new framework to improve customer service for citizens , accessed August 10, 2021.

    View in Article
  5. United States Department of Labor Office of Inspector General, “DOL–OIG oversight of the unemployment insurance program ,” June 10, 2021.

    View in Article
  6. William D. Eggers et al., Rebuilding trust in government: Four signals that can help improve citizen trust and engagement , Deloitte Insights, March 9, 2021.

    View in Article
  7. Megan Brenan, “Amid pandemic, confidence in key US institutions surges ,” Gallup, August 12, 2020.

    View in Article
  8. Dani Anguiano, “California’s wildfire hell: How 2020 became the state’s worst ever fire season ,” Guardian , December 30, 2020; Timothy Rooks, “West coast fires will cost US economy dearly ,” Deutsche Welle , September 17, 2020.

    View in Article
  9. Jacob Gershman, “A guide to state coronavirus reopenings and lockdowns ,” Wall Street Journal , May 20, 2020; Laura Hallas et al., Variation in US states’ COVID-19 policy responses , Blavatnik School of Government, May 7, 2021.

    View in Article

The authors would like to thank David Levin for inputs on the survey weighting and Kannan Thirumalai for his contributions to the survey data analysis.

The authors would also like to thank William EggersMahesh KelkarJesse GoldhammerSam KortaAmy FreckmannJoshua KnightJared Hall, and Greg Lewis for contributing their time and insights into the report.

Cover artwork: Kevin Weier.

Did you find this useful?

Thanks for your feedback

If you would like to help improve further, please complete a 3-minute survey